Two human dendritic cell (DC) subpopulations are recognized: myeloid (MDC) and plasmacytoid (PDC), which mediate distinct immunologic functions. Malignant ascites contained a striking accumulation of PDC, whereas MDC were entirely absent. Tumor draining lymph nodes (LN) contained both PDC and MDC although with marked phenotypic and functional differences compared to blood DC. Little regarding PDC mediated immunity is known. CD4+CD25+ regulatory T cells (Tregs) are elevated in human cancers where they inhibit TAA-specific immunity and predict poor survival. We recently demonstrated a role for PDC in immunopathogenesis of ovarian cancer, including induction of Tregs. We hypothesize that tumor PDC-induced Tregs inhibit tumor-associated antigen (TAA)-specific immunity and facilitate tumor growth, thus contributing to cancer immunopathogenesis. Tumor PDC express high-level B7-H1 that induces T cell IL-10. We will test the hypothesis that tumor PDC B7-H1 is a molecular mechanism for Treg differentiation. Factors inducing Treg differentiation are poorly understood. Understanding these processes is central to unraveling the immunopathologic basis of cancer and developing novel immune-based therapies. Our specific aims are: AIM 1: Test the hypothesis that tumor PDC-activated T cells block TAA-specific immunity. We will use well-characterized in vitro models for human immunity to test effects of naive T cell priming and on defined TAA-specific effector cell functions including cytokine secretion, proliferation, and cytolytic activity. We have preliminarily identified a PDC-activated CD8+ Treg which will be further studied. AIM 2: Test the hypothesis that PDC-activated T cells block immune-mediated tumor rejection in vivo. We developed novel chimeric SCID/NOD mouse models for these studies in which autologous human tumor, effector cells and Treg can be studied in vivo. This model allows testing of Treg function, TAA-specific immune suppression and tumor growth in vivo in an autologous setting. Mechanisms will be tested by blocking IL-10, CTLA-4 or other Treg effector arms in vivo.